Sunday, May 24, 2009

I'm a People Person!

Or I would be if I had greater gray matter density in my left and right inferior temporal lobes, my orbitofrontal cortex and ventral striatum (collapsed into one big region of interest), and my left putamen and pallidum. Plus lower gray matter density in my left and right cerebellum (Lebreton et al., 2009).


click figure for larger view
Fig. 1 (from Lebreton et al., 2009). Regions in which gray matter density (GMD) is associated with reward dependence. Mean gray matter density was extracted from each of the clusters that we identified using linear regression, transformed into Z-scores, and plotted versus RD. The lines represent the best fit for the associations when adjusted for total gray matter volume.

To be fair to the authors, they never used the expression "people person" in their paper. That was the BBC1, among other press outlets:
'People-person' brain area found

Scientists say they have located the brain areas that may determine how sociable a person is.

Warm, sentimental people tend to have more brain tissue in the outer strip of the brain just above the eyes and in a structure deep in the brain's centre.

These are the same zones that allow us to enjoy chocolate and sex, the Cambridge University experts report in the European Journal of Neuroscience.

. . .

The men who scored higher on questionnaire-based ratings of emotional warmth and sociability had more grey matter in two brain areas - the orbitofrontal cortex and ventral striatum.
As we can see in Figure 1, however, the gray matter volume in other brain areas (as measured using voxel-based morphometry) was also correlated with social reward dependence, defined as
a stable pattern of attitudes and behaviour hypothesized to represent a favourable disposition towards social relationships and attachment as a personality dimension.
Social reward dependence (RD) was measured using
the RD scale of Cloninger’s temperament and character inventory (TCI), a questionnaire that maps independent temperament traits onto putatively independent neurobiological systems (Cloninger, 1987, 1994). The RD subscale measures the tendency of subjects to be sensitive to a socially defined reward: a high score indicates a high disposition to social relationships and attachment; and a low score indicates a tendency to insensitivity and aloofness.
The participants were drawn from a large cohort of Finnish people born in 1966:
Data on biological, socioeconomic and health conditions, living habits, and family characteristics of cohort members were collected prospectively from pregnancy. The present study is based on 10,934 individuals living in Finland at the age of 16 years who did not forbid the use of their data...
Of that gigantic cohort, the authors restricted subject selection to 4,349 people (1,974 males) who had completed the TCI questionnaire. Of the 1,531 living in the city of Oulu, 187 subjects were randomly selected and invited to participate, and 104 (62 men) agreed to have an MRI. The authors excluded female participants from the present study because they're more emotional and different from men [basically]. And it turned out that TCI data were not available from 21 of the 62 men after all, so the final sample was comprised of 41 male volunteers.

Voxel-based morphometry was used to create probabilistic maps of gray matter for each individual. Associations between RD and brain structure were corrected for total gray matter volume and tested by fitting a multiple linear regression model at each voxel, followed by permutation testing. Out of the blue (with no previous or subsequent mention), the authors also decided to adjust for novelty seeking and harm avoidance. Lebreton et al. found the predicted correlation of RD and GMD in the orbitofrontal cortex (OFC) and the ventral striatum. They didn't have much to say about why the inferior temporal gyrus2 and cerebellum showed positive/negative correlations with RD, though. In the BBC article,

Professor Simon Baron Cohen, of the Autism Research Centre in Cambridge, said: "This is an important study in showing that the degree to which we find socializing rewarding is correlated with differences in brain structure.

"It reminds us that for some people, socializing is an intrinsic reward, just like chocolate or cannabis. And that what you find rewarding depends on differences in the brain.

Will we now see neurotraining programs designed to increase the size of your OFC and ventral striatum?


Footnotes

1 Found via @vaughanbell.

2 They called this region the temporal poles, but it looked more like ITG to me.

Reference

ResearchBlogging.org

Lebreton, M., Barnes, A., Miettunen, J., Peltonen, L., Ridler, K., Veijola, J., Tanskanen, P., Suckling, J., Jarvelin, M., Jones, P., Isohanni, M., Bullmore, E., & Murray, G. (2009). The brain structural disposition to social interaction. European Journal of Neuroscience DOI: 10.1111/j.1460-9568.2009.06782.x.

Subscribe to Post Comments [Atom]

5 Comments:

At May 25, 2009 12:33 AM, Blogger Neuroskeptic said...

Are those scatterplots products of the Vul et al error? I guess not if they're based on anatomical ROIs...

 
At May 25, 2009 8:28 AM, Anonymous Anonymous said...

One wonders if the findings also hold for Dog Persons.

 
At May 25, 2009 4:04 PM, Blogger The Neurocritic said...

Anonymous - Ha, ha! It turns out that dog experts have been studied in the context of the debate on human face processing abilities as innate vs. acquired expertise (Robbins & McKone, 2007).

Neuroskeptic - Lebreton et al. didn't use ROIs, they did a whole brain analysis to create probabilistic maps of gray matter, white matter and CSF for each subject. Then they used non-parametric statistics (permutation testing) to test for significant correlations:

"Associations between RD and adult brain structure, adjusted for total gray matter volume, novelty seeking and harm avoidance were tested by multiple linear regression onto the values of GMD at each voxel using permutation-based methods implemented in Cambridge Brain Analysis (CamBA) software..."

According to the Vul and Kanwisher chapter (PDF):

"Permutation tests are one particularly effective method when analyses are particularly complicated. Researchers can randomly permute the condition labels for their data and undertake the same analysis. If this is done enough times, it is possible to empirically estimate the probability of the outcome observed with the true data labels. Unlike simpler (and faster) white-noise simulations, this permutation analysis includes the non-gaussian structure of the BOLD signal noise, and is thus more accurate."

 
At May 26, 2009 4:52 PM, Anonymous Anonymous said...

It is interesting that the Vul & al. paper has simply generated suspicion for scatterplots of this kind without generating real understanding of the problem (in part, because neither of them are statisticians and because they hyped the problem for their Blair Witch Project style pre-release campaign).
If you correct for multiple comparisons you are fine, for correlations and for simple effects (you'll get your 5% of false positives, but that is life). The question is how to correct for multiple comparisons. If you use Bonferroni, you are OK by definition, it's a worst case scenario. A less extreme scenario models the spatial correlation in the data. Monte Carlo methods work well in this case. Also, if your ROIs include voxels that are all above the correlation threshold, then presenting a scatterplot of the average within the ROI is usually not problematic.
All statistics are surrogates for replication. Our field desperately needs a Journal of Null Effects.

 
At June 01, 2009 1:13 AM, Blogger Neuroskeptic said...

Anon: I'm not suspicious of all scatterplots, but Vul et al correctly point out that a scatterplot of this nature is essentially meaningless if the voxels across which the plot is calculated (which is often not stated precisely) are chosen in certain ways.

It's not true that "If you correct for multiple comparisons you are fine". If you correct for multiple comparisons, you only have a 5% chance of getting a result out of pure noise.

But, and this is Vul's point, you can create an amazingly impressive result (correlation coefficent of .8 or more, and a very pretty scatterplot) out of a very weak one, by doing a non-independent analysis.

 

Post a Comment

Links to this post:

Create a Link

<< Home

eXTReMe Tracker